OpenAI and Broadcom unveil Jalapeño, OpenAI's first custom LLM inference chip
On 2026-06-24 OpenAI and Broadcom jointly unveiled Jalapeño, OpenAI's first custom Intelligence Processor, with engineering samples running ML workloads in the lab at production target frequency and power and a multi-generation platform described as gigawatt-scale with Microsoft and other partners beginning in 2026. Per-watt performance, the 9-month tape-out timeline, and a 'GPT-5.3-Codex-Spark' workload are vendor claims; no independent benchmark, no shipping date, and no SKU has been published.